A Neural Network for Nonlinear Optimization with General Linear Constraints

In this study, we investigate a novel neural network for solving nonlinear convex programming problems with general linear constraints. Furthermore, we extend this neural network to solve a class of variational inequalities problems. These neural networks are stable in the sense of Lyapunov and globally convergent to a unique optimal solution. The present convergence results do not requires Lipschitz continuity condition on the objective function. These models have no adjustable parameter and have a low complexity for implementation and converge to an exact optimal solution.